Package dev.nm.stat.timeseries.linear.multivariate.stationaryprocess.arma

Class VARMAAutoCovariance

java.lang.Object

dev.nm.analysis.function.matrix.R2toMatrix

dev.nm.stat.timeseries.linear.multivariate.MultivariateAutoCovarianceFunction

dev.nm.stat.timeseries.linear.multivariate.stationaryprocess.arma.VARMAAutoCovariance

All Implemented Interfaces:

Function<Vector,Matrix>, RntoMatrix

public class VARMAAutoCovariance
extends MultivariateAutoCovarianceFunction

Compute the Auto-CoVariance Function (ACVF) for a vector AutoRegressive Moving Average (ARMA) model, assuming that EX_t = 0.

This implementation solves the Yule-Walker equation.

The R equivalent functions are ARMAacf and TacvfAR in package FitAR.

Nested Class Summary

Nested classes/interfaces inherited from interface dev.nm.analysis.function.Function

Function.EvaluationException

Constructor Summary

Constructors

Constructor	Description
VARMAAutoCovariance(VARMAModel model, int nLags)	Compute the auto-covariance function for a vector ARMA model.

Method Summary

Modifier and Type	Method	Description
Matrix	evaluate(double i)	Get the i-th auto-covariance matrix.
Matrix	evaluate(double i, double j)	Evaluate f(x1, x2) = A.

Methods inherited from class dev.nm.stat.timeseries.linear.multivariate.MultivariateAutoCovarianceFunction

get

Methods inherited from class dev.nm.analysis.function.matrix.R2toMatrix

dimensionOfDomain, dimensionOfRange, evaluate

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

VARMAAutoCovariance

public VARMAAutoCovariance(VARMAModel model,
 int nLags)

Compute the auto-covariance function for a vector ARMA model.

To solve Eq. 11.3.15, we "expand" the (p+1) matrix equations into (p+1)*m*m linear equations. m is the dimension of Γ (ACVF).

Parameters:

model - an ARIMA model

nLags - the number of lags

Method Details

evaluate

public Matrix evaluate(double i,
 double j)

Description copied from class: R2toMatrix

Evaluate f(x₁, x₂) = A.

Specified by:

evaluate in class R2toMatrix

Parameters:

i - x1

j - x2

Returns:

f(x1, x2)

evaluate

public Matrix evaluate(double i)

Get the i-th auto-covariance matrix.

Parameters:

i - the lag order

Returns:

the i-th auto-covariance matrix